Source: http://www1.ics.uci.edu/pub/machine-learning-databases/space-shuttle/

1. Title: Challenger Space Shuttle O-Ring Data (2 databases)

2. Sources:
   -- David Draper (draper@math.ucla.edu)
      University of California, Los Angeles
   -- Donor: David Draper (draper@math.ucla.edu)
   -- Date: 5 August 1993

3. Past Usage:

   1. Draper,~D. (1993).  Assessment and propagation of model uncertainty.  
      In {\it Proceedings of the Fourth International Workshop on Artificial
      Intelligence and Statistics} (pp. 497--509).  Ft. Lauderdale, FL:  
      Unpublished.
      -- Discrete model uncertainty analysis
      -- Analysis suggests that obvious different extrapolations of the
         data exist at 31 degrees Fahrenheit (i.e., freezing), which sharply
         discredits the assumption of no temperature effect.
   2. Dalal,~S.~R., Fowlkes,~E.~B., \& Hoadley,~B. (1989). Risk analysis of
      the space shuttle: pre-Challenger prediction of failure. {\it Journal
      of the American Statisticians Association}, {\it 84}, 945--957.
   3. Lavine,~M. (1991). Problems in extrapolation illustrated with space 
      shuttle O-ring data.  {\it Journal of the American Statisticians
      Association}, {\it 86}, 919--922.
   4. Martz~H.~F., \& Zimmer,~W.~J. (1992). The risk of catastrophic failure
      of the solid rocket boosters on the space shuttle.  {\it American
      Statistics}, {\it 46}, 42--47. 

4. Number of instances: 23 in each of two files

5. Relevant Information:

   There are two databases: (both use the same set of 5 attributes)
     1. Primary o-ring erosion and/or blowby
     2. Primary o-ring erosion only
   The two databases are identical except for the 2nd attribute of the
   21st instance (confirmed by David Draper on 8/5/93).

   Edited from (Draper, 1993):
      The motivation for collecting this database was the explosion of the
   USA Space Shuttle Challenger on 28 January, 1986.  An investigation
   ensued into the reliability of the shuttle's propulsion system.  The
   explosion was eventually traced to the failure of one of the three field 
   joints on one of the two solid booster rockets.  Each of these six field 
   joints includes two O-rings, designated as primary and secondary, which
   fail when phenomena called erosion and blowby both occur. 
      The night before the launch a decision had to be made regarding
   launch safety.  The discussion among engineers and managers leading to
   this decision included concern that the probability of failure of the
   O-rings depended on the temperature t at launch, which was forecase to
   be 31 degrees F. There are strong engineering reasons based on the
   composition of O-rings to support the judgment that failure
   probability may rise monotonically as temperature drops.  One other
   variable, the pressure s at which safety testing for field join leaks
   was performed, was available, but its relevance to the failure process
   was unclear.
       Draper's paper includes a menacing figure graphing the number of field
   joints experiencing stress vs. liftoff temperature for the 23 shuttle 
   flights previous to the Challenger disaster.  No previous liftoff 
   temperature was under 53 degrees F.  Although tremendous extrapolation
   must be done from the given data to assess risk at 31 degrees F, it
   is obvious even to the layman "to foresee the unacceptably high risk
   created by launching at 31 degrees F."  For more information, see
   Draper (1993) or the other previous analyses.
       The task is to predict the number of O-rings that will experience
   thermal distress for a given flight when the launch temperature is 
   below freezing.

6. Number of Attributes: 5
     1. Number of O-rings at risk on a given flight
     2. Number experiencing thermal distress
     3. Launch temperature (degrees F)
     4. Leak-check pressure (psi)
     5. Temporal order of flight

7. Attribute Information: all values are positive integers